
## List of files

1. 'final-many-to-many.csv': This is the main replication data file. It includes country-year estimates of mass-elite congruence by economic class background and a number of covariates. Note that there are a number of covariates in the data that were used in previous iterations of this paper but are no longer used.
2. 'partyweights.csv': This is a file used in the construction of our data; it provides our hand-coded partisan weights used for post-stratifying legislator responses as described in the text. (The gender weights are a covariate 'pc_women_elite_real' in the 'final-many-to-many.csv' dataset.)
3. 'Lupu-Warner-Affluence-data-cleaning-replication.R': This is the script that cleans and merges all of the survey data and constructs all of the aggregated data used in the analysis.
4. 'Lupu-Warner-Affluence-replication.R': This is the script that replicates all of the findings in the paper.
5. 'Lupu-Warner-Affluence-Appendix-replication': This is the script that replicates all of the findings in the online appendix to the paper which are not produced via the 'Lupu-Warner-Affluence-replication.R' script.


## A note on data access

Unfortunately, data use agreements prohibit us from sharing the survey data on which our results are based. Instead, we provide the code used to clean and merge these data in the data cleaning script. Please find the list of data sources and notes on availability in the online appendix.

We have made every effort to provide the code in a format that is legible and organized, but please do email us if you're encountering difficulty reproducing our results.